FLAMENCO Learning Disabilities Dataset

In the context of the FLAMENCO project, we have released a dataset designed for predicting potential deficiencies in children's communication skills, tailored for Federated Learning. This dataset specifically focuses on addressing two prevalent deficiencies in communication skill development in children: autism and intellectual disability. For each deficiency, two CSV files are provided—one for training machine learning models and another for testing them. Each entry in these CSV files includes the following details:

-        case_id: An anonymized identifier used to distinguish cases.

-        client_id: Identifies the client to which the case belongs, useful for dataset splitting in federated settings.

-        A series of scores measuring specific communication skills:  These scores, such as Verbalization, Voicing, Syntax, etc., are derived from the child's performance in specialised gamified exercises and have been computed with the assistance of expert clinicians.

        target: Can be -1 (no clinician's diagnosis available for the case), 0 (no diagnosed deficiency in the case), 1 (indicates a positive diagnosis of communication deficiency by a clinician)

TThis project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreements No. 957406 (TERMINET).

在FLAMENCO项目框架下，我们发布了一款专为联合学习（Federated Learning）设计的儿童沟通技能潜在缺陷预测数据集。
该数据集聚焦于儿童沟通技能发展中两类高发缺陷：自闭症（autism）与智力障碍（intellectual disability）。针对每类缺陷，我们提供了两份CSV文件——一份用于机器学习模型训练，另一份用于模型测试。

上述CSV文件中的每条样本均包含以下详细信息：
- case_id：用于区分样本的匿名标识符
- client_id：标识该样本所属的客户端，便于联合学习场景下的数据集划分
- 一系列特定沟通技能评分：诸如语言表达、发声、句法结构等评分，均源自儿童在专业化游戏化练习中的表现，并由临床专家协助计算得出。

目标标签（target）的取值可为：-1（该样本无临床医生诊断结果）、0（该样本未被诊断出沟通缺陷）、1（临床医生确诊该样本存在沟通缺陷）

本项目获得欧盟地平线2020研究与创新计划资助，资助协议编号为957406（TERMINET）。